Github user StephanEwen commented on the pull request:

    https://github.com/apache/flink/pull/1214#issuecomment-145818892
  
    Can we step back a bit with this issue? The design for this seems quite 
crucial to get right to not get extreme code complexity/bloat and make 
everything about job submission, accumulators, etc very hard in the future.
    
    What is the design behind this pull request?
      - Can the "detached mode" handle interactive programs?
      - What about programs that use accumulators?
      - If not, how would they be recognized and what errors can be given?
      - The problem here seems very similar to intercepting the execution plan 
of a job and returning it. Any code sharing?


---
If your project is set up for it, you can reply to this email and have your
reply appear on GitHub as well. If your project does not have this feature
enabled and wishes so, or if the feature is enabled but not working, please
contact infrastructure at [email protected] or file a JIRA ticket
with INFRA.
---

Reply via email to